Rearrangements

lib/LaTeXML/Common/Font.pm

@@ -60,44 +60,53 @@ my $FLAG_EMPH         = 0x10;
 # NOTE: This probably doesn't really belong in here...
 
 my %font_family = (
-  cmr   => { family => 'serif' },      cmss  => { family => 'sansserif' },
-  cmtt  => { family => 'typewriter' }, cmvtt => { family => 'typewriter' },
-  cmt   => { family => 'serif' },    # for cmti "text italic"
-  cmfib => { family => 'serif' },      cmfr  => { family => 'serif' },
-  cmdh  => { family => 'serif' },      cm    => { family => 'serif' },
-  ptm   => { family => 'serif' },      ppl   => { family => 'serif' },
-  pnc   => { family => 'serif' },      pbk   => { family => 'serif' },
-  phv   => { family => 'sansserif' },  pag   => { family => 'serif' },
-  pcr   => { family => 'typewriter' }, pzc   => { family => 'script' },
-  put   => { family => 'serif' },      bch   => { family => 'serif' },
-  psy   => { family => 'symbol' },     pzd   => { family => 'dingbats' },
-  ccr   => { family => 'serif' },      ccy   => { family => 'symbol' },
-  cmbr  => { family => 'sansserif' },  cmtl  => { family => 'typewriter' },
-  cmbrs => { family => 'symbol' },     ul9   => { family => 'typewriter' },
-  txr   => { family => 'serif' },      txss  => { family => 'sansserif' },
-  txtt  => { family => 'typewriter' }, txms  => { family => 'symbol' },
-  txsya => { family => 'symbol' },     txsyb => { family => 'symbol' },
-  pxr   => { family => 'serif' },      pxms  => { family => 'symbol' },
-  pxsya => { family => 'symbol' },     pxsyb => { family => 'symbol' },
-  futs  => { family => 'serif' },
-  uaq   => { family => 'serif' },   ugq  => { family => 'sansserif' },
-  eur   => { family => 'serif' },   eus  => { family => 'script' },
-  euf   => { family => 'fraktur' }, euex => { family => 'symbol' },
+  cmr    => { family => 'serif' },
+  cmss   => { family => 'sansserif' },
+  cmssq  => { family => 'sansserif' },                       # quote style?
+  cmssqi => { family => 'sansserif', shape => 'italic' },    # quote style?
+  cmtt   => { family => 'typewriter' }, cmvtt => { family => 'typewriter' },
+  cmt    => { family => 'serif' },                           # for cmti "text italic"
+  cmfib  => { family => 'serif' },
+  cmfr   => { family => 'serif' },
+  cm     => { family => 'serif' },
+  cmdh   => { family => 'serif' },
+  cmr    => { family => 'serif' },
+  cmdunh => { family => 'serif' },                           # like cmr10 but with tall body heights
+  cmu    => { family => 'serif' },                           # unslanted italic ??
+  ptm    => { family => 'serif' },      ppl   => { family => 'serif' },
+  pnc    => { family => 'serif' },      pbk   => { family => 'serif' },
+  phv    => { family => 'sansserif' },  pag   => { family => 'serif' },
+  pcr    => { family => 'typewriter' }, pzc   => { family => 'script' },
+  put    => { family => 'serif' },      bch   => { family => 'serif' },
+  psy    => { family => 'symbol' },     pzd   => { family => 'dingbats' },
+  ccr    => { family => 'serif' },      ccy   => { family => 'symbol' },
+  cmbr   => { family => 'sansserif' },  cmtl  => { family => 'typewriter' },
+  cmbrs  => { family => 'symbol' },     ul9   => { family => 'typewriter' },
+  txr    => { family => 'serif' },      txss  => { family => 'sansserif' },
+  txtt   => { family => 'typewriter' }, txms  => { family => 'symbol' },
+  txsya  => { family => 'symbol' },     txsyb => { family => 'symbol' },
+  pxr    => { family => 'serif' },      pxms  => { family => 'symbol' },
+  pxsya  => { family => 'symbol' },     pxsyb => { family => 'symbol' },
+  futs   => { family => 'serif' },
+  uaq    => { family => 'serif' },   ugq  => { family => 'sansserif' },
+  eur    => { family => 'serif' },   eus  => { family => 'script' },
+  euf    => { family => 'fraktur' }, euex => { family => 'symbol' },
   # The following are actually math fonts.
-  ms    => { family => 'symbol' },
-  ccm   => { family => 'serif',      shape    => 'italic' },
-  cmm   => { family => 'italic',     encoding => 'OML' },
-  cmex  => { family => 'symbol',     encoding => 'OMX' },      # Not really symbol, but...
-  cmsy  => { family => 'symbol',     encoding => 'OMS' },
-  ccitt => { family => 'typewriter', shape    => 'italic' },
-  cmbrm => { family => 'sansserif',  shape    => 'italic' },
-  futm  => { family => 'serif',      shape    => 'italic' },
-  futmi => { family => 'serif',      shape    => 'italic' },
-  txmi  => { family => 'serif',      shape    => 'italic' },
-  pxmi  => { family => 'serif',      shape    => 'italic' },
-  bbm   => { family => 'blackboard' },
-  bbold => { family => 'blackboard' },
-  bbmss => { family => 'blackboard' },
+  ms     => { family => 'symbol' },
+  ccm    => { family => 'serif',      shape    => 'italic' },
+  cmm    => { family => 'math',       shape    => 'italic', encoding => 'OML' },
+  cmex   => { family => 'symbol',     encoding => 'OMX' },       # Not really symbol, but...
+  cmsy   => { family => 'symbol',     encoding => 'OMS' },
+  ccitt  => { family => 'typewriter', shape    => 'italic' },
+  cmsltt => { family => 'typewriter', shape    => 'slanted' },
+  cmbrm  => { family => 'sansserif',  shape    => 'italic' },
+  futm   => { family => 'serif',      shape    => 'italic' },
+  futmi  => { family => 'serif',      shape    => 'italic' },
+  txmi   => { family => 'serif',      shape    => 'italic' },
+  pxmi   => { family => 'serif',      shape    => 'italic' },
+  bbm    => { family => 'blackboard' },
+  bbold  => { family => 'blackboard' },
+  bbmss  => { family => 'blackboard' },
   # some ams fonts
   cmmib => { family => 'italic', series   => 'bold' },
   cmbsy => { family => 'symbol', series   => 'bold' },
@@ -177,16 +186,17 @@ sub decodeFontname {
     if (my $ffam = lookupFontFamily($fam)) { map { $props{$_} = $$ffam{$_} } keys %$ffam; }
     if (my $fser = lookupFontSeries($ser)) { map { $props{$_} = $$fser{$_} } keys %$fser; }
     if (my $fsh  = lookupFontShape($shp))  { map { $props{$_} = $$fsh{$_} } keys %$fsh; }
-    $size = 1 unless $size;    # Yes, also if 0, "" (from regexp)
-    $size = $at if defined $at;
-    $size *= $scaled if defined $scaled;
+    $size        = 1 unless $size;    # Yes, also if 0, "" (from regexp)
+    $size        = $at             if defined $at;
+    $size        = $size * $scaled if defined $scaled;
+    $props{name} = $name;
     $props{size} = $size;
     # Experimental Hack !?!?!?
     $props{encoding} = 'OT1' unless defined $props{encoding};
-    $props{at}       = $at . "pt" if defined $at;
     return %props; }
   else {
-    return; } }
+    Info('unrecognized', 'font', undef, "Unrecognized fontname '$name'");
+    return (family => $name, size => DEFSIZE()); } }
 
 sub lookupTeXFont {
   my ($fontname, $seriescode, $shapecode) = @_;
@@ -266,6 +276,7 @@ sub stringify {
   no warnings 'recursion';
   my ($self) = @_;
   my ($fam, $ser, $shp, $siz, $col, $bkg, $opa, $enc, $lang, $mstyle, $flags) = @$self;
+  # !!!!!
   $fam = 'serif' if $fam && ($fam eq 'math');
   return 'Font[' . join(',', map { Stringify($_) } grep { $_ }
       (isDiff($fam, $DEFFAMILY)   ? ($fam)    : ()),
@@ -280,6 +291,16 @@ sub stringify {
     )
     . ']'; }
 
+# Return a Fontinfo-like hash
+# Eventually a more integrated representation of Fonts that accommodates
+# both low-level TeX-like commands, and higher-level CSS-like ones.
+sub asFontinfo {
+  my ($self) = @_;
+  my ($fam, $ser, $shp, $siz, $col, $bkg, $opa, $enc, $lang, $mstyle, $flags) = @$self;
+  return { family => $fam, series => $ser, shape => $shp, size => $siz,
+    color    => $col,          background => $bkg,  opacity   => $opa,
+    encoding => $enc || 'OT1', language   => $lang, mathstyle => $mstyle }; }
+
 sub equals {
   my ($self, $other) = @_;
   return (defined $other) && ((ref $self) eq (ref $other))
@@ -328,6 +349,7 @@ sub relativeTo {
   my ($self, $other) = @_;
   my ($fam,  $ser,  $shp,  $siz,  $col,  $bkg,  $opa,  $enc,  $lang,  $mstyle,  $flags)  = @$self;
   my ($ofam, $oser, $oshp, $osiz, $ocol, $obkg, $oopa, $oenc, $olang, $omstyle, $oflags) = @$other;
+  # !!!!
   $fam  = 'serif' if $fam  && ($fam eq 'math');
   $ofam = 'serif' if $ofam && ($ofam eq 'math');
 ##  my $emph = 0;
@@ -358,6 +380,9 @@ sub relativeTo {
     (isDiff($opa, $oopa)
       ? (opacity => { value => $opa, properties => { opacity => $opa } })
       : ()),
+    (isDiff($enc, $oenc)
+      ? (encoding => { value => $enc, properties => { encoding => $enc } })
+      : ()),
     (isDiff($lang, $olang)
       ? ('xml:lang' => { value => $lang, properties => { language => $lang } })
       : ()),

lib/LaTeXML/Core/Definition/CharDef.pm

@@ -47,13 +47,17 @@ sub invoke {
   my $mathglyph = $$self{mathglyph};
   # A dilemma: If the \chardef were in a style file, you're prefer to revert to the $cs
   # but if defined in the document source, better to use \char ###\relax, so it still "works"
-  if (defined $mathglyph) {    # Must be a math char
+  my $src   = $$self{locator} && $$self{locator}->toString;
+  my $local = $src && $src !~ /\.(?:sty|ltxml|ltxmlc)/;    # Dumps currently have undefined src!
+  if (defined $mathglyph) {                                # Must be a math char
     return Box($mathglyph, undef, undef,
-      Tokens(T_CS('\mathchar'), $value->revert, T_CS('\relax')),
+      ($local ? Tokens(T_CS('\mathchar'), $value->revert, T_CS('\relax')) : $$self{cs}),
       role => $$self{role}); }
-  else {                       # else text; but note defered font/encoding till digestion!
-    return Box(LaTeXML::Package::FontDecode($value->valueOf), undef, undef,
-      Tokens(T_CS('\char'), $value->revert, T_CS('\relax'))); } }
+  else {    # else text; but note defered font/encoding till digestion!
+    my ($char, %props) = LaTeXML::Package::FontDecode($value->valueOf);
+    return Box($char, undef, undef,
+      ($local ? Tokens(T_CS('\char'), $value->revert, T_CS('\relax')) : $$self{cs}),
+      %props); } }
 
 sub equals {
   my ($self, $other) = @_;

lib/LaTeXML/Core/State.pm

@@ -124,13 +124,12 @@ sub new {
   $$self{delcode}             = {};
   $$self{tracing_definitions} = {};
   # Initializations that INITEX would have set.
-  $$self{mathcode}{'.'} = [0];
   for (my $c = ord('0') ; $c <= ord('9') ; $c++) {
-    $$self{mathcode}{ chr($c) } = [0x7000]; }
+    $$self{mathcode}{ chr($c) } = [0x7000 + $c]; }
   for (my $c = ord('a') ; $c <= ord('z') ; $c++) {
     my $C = $c + ord('A') - ord('a');
-    $$self{mathcode}{ chr($c) } = [0x7100];
-    $$self{mathcode}{ chr($C) } = [0x7100];
+    $$self{mathcode}{ chr($c) } = [0x7100 + $c];
+    $$self{mathcode}{ chr($C) } = [0x7100 + $C];
     $$self{uccode}{ chr($c) }   = [$C];
     $$self{lccode}{ chr($C) }   = [$c];
     $$self{sfcode}{ chr($C) }   = [999]; }

lib/LaTeXML/Engine/LaTeX.pool.ltxml

@@ -4889,6 +4889,7 @@ DefConstructor('\@framebox[Dimension][]{}',
           $document->setAttribute($c[0], $k => $v); } } } }
 );
 
+AssignValue(allocated_boxes => 0, 'global');
 DefPrimitive('\newsavebox DefToken', sub {
     my $n = LookupValue('allocated_boxes') + 1;
     AssignValue(allocated_boxes => $n, 'global');

lib/LaTeXML/Engine/TeX_Character.pool.ltxml

@@ -185,7 +185,7 @@ DefRegister('\catcode Number', Number(0),
 # Not used anywhere (yet)
 DefRegister('\sfcode Number', Number(0),
   getter => sub { my $code = $STATE->lookupSFcode(chr($_[0]->valueOf));
-    Number(defined $code ? $code : 0); },
+    Number(defined $code ? $code : 1000); },
   setter => sub { $STATE->assignSFcode(chr($_[2]->valueOf) => $_[0]->valueOf, $_[1]); });
 DefRegister('\lccode Number', Number(0),
   getter => sub { my $code = $STATE->lookupLCcode(chr($_[0]->valueOf));

lib/LaTeXML/Engine/TeX_Fonts.pool.ltxml

@@ -285,5 +285,38 @@ DeclareFontMap('OMX',
     #                        [missing tips for horizontal curly braces]
     "\x{2191}", "\x{2193}", undef, undef, undef, undef, "\x{21D1}", "\x{21D3}"]);
 
+#%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+# TeX's ligatures handled by rewrite regexps.
+# Note: applied in reverse order of definition (latest defined applied first!)
+# Note also, these area only applied in text content, not in attributes!
+sub nonTypewriter {
+  my ($font) = @_;
+  return ($font->getFamily ne 'typewriter'); }
+
+sub nonTypewriterT1 {
+  my ($font) = @_;
+  return ($font->getFamily ne 'typewriter') && (($font->getEncoding || 'OT1') =~ /^(OT1|T1)$/); }
+
+# EN DASH (NOTE: With digits before & aft => \N{FIGURE DASH})
+DefLigature(qr{--},  "\x{2013}", fontTest => \&nonTypewriter);    # EN dash
+DefLigature(qr{---}, "\x{2014}", fontTest => \&nonTypewriter);    # EM dash
+
+# Ligatures for doubled single left & right quotes to convert to double quotes
+# [should ligatures be part of a font, in the first place? (it is in TeX!)
+DefLigature(qr{\x{2018}\x{2018}}, "\x{201C}", fontTest => \&nonTypewriterT1);   # double left quote
+DefLigature(qr{\x{2019}\x{2019}}, "\x{201D}", fontTest => \&nonTypewriterT1);   # double right quote
+DefLigature(qr{\?\x{2018}},       UTF(0xBF),  fontTest => \&nonTypewriterT1);   # ? backquote
+DefLigature(qr{!\x{2018}},        UTF(0xA1),  fontTest => \&nonTypewriterT1);   # ! backquote
+# These ligatures are also handled by TeX.
+# However, it appears that decent modern fonts in modern browsers handle these at that level.
+# So it's likely not worth doing it at the conversion level, possibly adversely affecting search.
+# DefLigature(qr{ff},               "\x{FB00}", fontTest => \&nonTypewriterT1);
+# DefLigature(qr{fi},               "\x{FB01}", fontTest => \&nonTypewriterT1);
+# DefLigature(qr{fl},               "\x{FB02}", fontTest => \&nonTypewriterT1);
+# DefLigature(qr{ffi},              "\x{FB03}", fontTest => \&nonTypewriterT1);
+# DefLigature(qr{ffl},              "\x{FB04}", fontTest => \&nonTypewriterT1);
+
+DefLigature(qr{\.\.\.}, "\x{2026}", fontTest => \&nonTypewriter);    # ldots
+
 #%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 1;

lib/LaTeXML/Engine/TeX_Math.pool.ltxml

@@ -1193,5 +1193,14 @@ DefConstructor('\lx@eqno{}',
   "^ <ltx:tags><ltx:tag><ltx:Math><ltx:XMath>#1</ltx:XMath></ltx:Math></ltx:tag></ltx:tags>",
   reversion => '');
 
+#======================================================================
+# Pretest for XMath to keep from interpreting math that the DOM may not allow!!
+##DefMathRewrite(xpath=>'descendant-or-self::ltx:XMath',match=>'\cdot\cdot\cdot',replace=>'\cdots');
+
+DefMathLigature("\x{22C5}\x{22C5}\x{22C5}" => "\x{22EF}", role => 'ID', name => 'cdots');
+
+#DefMathRewrite(xpath=>'descendant-or-self::ltx:XMath',match=>'...',replace=>'\ldots');
+DefMathLigature("..." => "\x{2026}", role => 'ID', name => 'ldots');
+
 #%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 1;

lib/LaTeXML/Engine/TeX_Paragraph.pool.ltxml

@@ -68,6 +68,19 @@ DefConstructorI('\noindent', undef, sub {
     # Otherwise ignore.
     return; });
 
+sub alignLine {
+  my ($document, $line, $alignment) = @_;
+  if ($document->isOpenable('ltx:p')) {
+    $document->insertElement('ltx:p', $line, class => 'ltx_align_' . $alignment); }
+  elsif ($document->isOpenable('ltx:text')) {
+    $document->insertElement('ltx:text', $line, class => 'ltx_align_' . $alignment);
+    $document->insertElement('ltx:break'); }
+  else {
+    Info('unexpected', 'alignment', $document,
+      "Lost requested alignment '$alignment'; no suitable element");
+    $document->absorb($line); }
+  return; }
+
 # <ltx:para> represents a Logical Paragraph, whereas <ltx:p> is a `physical paragraph'.
 # A para can contain both p and displayed equations and such.